Three-class Markovian segmentation of high-resolution sonar images

This paper presents an original method for analyzing, in an unsupervised wa y, images supplied by high resolution sonar, We aim at segmenting the sonar image into three kinds of regions: echo areas (due to the reflection of th e acoustic wave on the object), shadow areas (corresponding to a lack of ac oustic reverberation behind an object lying on the sea-bed), and sea-bottom reverberation areas. This unsupervised method estimates the parameters of noise distributions, modeled by a Weibull probability density function (PDF ), and the label field parameters, modeled by a Markov random field (MRF), For the estimation step, we adopt a maximum likelihood technique for the no ise model parameters and a least-squares method to estimate the MRF prior m odel. Then, in order to obtain an accurate segmentation map, we have design ed a two-step process that finds the shadow and the echo regions separately , using the previously estimated parameters. First, we introduce a scale-ca usal and spatial model called SCM (scale causal multigrid), based on a mult igrid energy minimization strategy, to find the shadow class. Second, we pr opose a MRF monoscale model using a priori information (at different level of knowledge) based on physical properties of each region, which allows us to distinguish echo areas from sea-bottom reverberation. This technique has been successfully applied to real sonar images and is compatible with auto matic processing of massive amounts of data. (C) 1999 Academic Press.